The question "Dext or a spreadsheet?" usually gets asked by two very different people, and they deserve different answers.
The first is a bookkeeper running thirty client files, publishing into Xero or QuickBooks, and trying to stop supplier invoices arriving as photographs in WhatsApp. The second is a sole trader whose accountant mentioned Dext once, who files a single Self Assessment return a year, and who currently keeps everything in a folder called "receipts 2026".
For the first, Dext is usually the right tool and the rest of this piece will not change that. For the second, it is frequently more machinery than the job requires. The interesting part is where the line falls.
What Dext is actually built to do
Dext is practice software. Its design assumptions are visible in its feature set: many clients, many documents per client, documents arriving through multiple channels, and an end state of coded transactions sitting in an accounting ledger.
Its genuine strengths follow from that:
- Bulk document handling across clients. Capture by app, email-in address, or direct supplier fetch, with documents routed to the right client file. At volume this is the whole game, and a spreadsheet workflow has no answer to it.
- Supplier rules that compound. Teach it once how a given supplier's invoices should be coded, and subsequent invoices from that supplier arrive pre-coded. Across hundreds of recurring suppliers this removes a large fraction of manual work permanently.
- Publishing into the ledger. Transactions land in Xero, QuickBooks, Sage, or similar, with the source document attached. For anyone whose deliverable is a set of books, this is the step that matters and it is the one a spreadsheet cannot do.
- Fetching invoices automatically. Pulling statements and invoices directly from suppliers and utilities removes the chasing, which is often the most irritating part of the month.
- Practice workflow. Client lists, staff assignment, and status tracking — the administrative layer that becomes essential past a handful of clients and is invisible overhead below that.
If your work involves the phrase "my clients", that list describes real daily value and the subscription is straightforwardly justified.
What a Google Sheets workflow is actually good at
A spreadsheet workflow is a different proposition: receipts are scanned, the data lands as rows in a sheet you control, and the sheet is the record.
- The output is yours and it is portable. No export step, no vendor format, no question about what happens to your history if you stop paying. A CSV is readable in thirty years.
- The structure bends to the job. Property investors want a property column; tradies want a job column; e-commerce sellers want landed cost per SKU. In a spreadsheet those are columns you add in ten seconds. In structured accounting software they are either a supported dimension or they are not.
- Nobody needs training. Handing a sheet to a client who already uses spreadsheets involves no onboarding call. This matters more than it sounds for the clients who resist any new system.
- The cost is flat and small. Relevant when the alternative is per-client or per-document pricing on a very low document count.
- Nothing to configure before it works. Scan, row appears. There is no chart of accounts to map first.
The honest limits are just as clear. A sheet does not publish into a ledger. It has no supplier rules, no practice dashboard, no automatic invoice fetching, and no multi-client routing. At ten clients it would be a bad idea; at a hundred it is not a serious proposal.
Side by side
| Criterion | Dext | Google Sheets workflow |
|---|---|---|
| Many clients at volume | Built for it | Not viable past a few |
| Publishing to Xero/QuickBooks/Sage | Core function | Not supported |
| Supplier-level coding rules | Yes, and they compound | Manual, or basic formula matching |
| Automatic supplier invoice fetch | Yes | No |
| Custom dimensions (job, property, SKU) | Within its model | Add a column |
| Data portability | Export available | Native — the sheet is the data |
| Setup before first use | Account and ledger mapping | Connect a sheet |
| Cost shape | Practice/plan pricing | Flat, low |
| Client onboarding effort | Real | Near zero for spreadsheet users |
Where the line actually falls
Three questions settle it in most cases.
Does your deliverable end in an accounting ledger? If the job is not finished until transactions are posted in Xero, choose the tool that posts to Xero. Everything else is secondary.
How many suppliers recur? Supplier rules are the compounding advantage, and they only compound if the same suppliers appear month after month. A business with forty recurring suppliers gets enormous value. A consultant with twelve transactions a month, mostly one-off, gets almost none.
Who reads the output? If the answer is "my accountant, once a year", the output is a list of categorised transactions with evidence attached — a job a well-structured sheet does completely.
The case this piece exists for
There is a specific situation where the spreadsheet workflow wins clearly, and it is common: a sole trader or micro-business whose accountant uses practice software, but who personally needs nothing more than a clean record of what was spent and why.
In that arrangement the accountant's tooling is the accountant's concern. What the client needs to supply is a categorised list with legible evidence, delivered in a format that does not require the accountant to re-key anything. A Google Sheet with date, merchant, category, amount, VAT, and a link to each receipt image satisfies that completely, at a cost of a few pounds a month and no training.
This is where Scan2Sheet fits: it scans the receipt, extracts the fields, categorises the row, flags a duplicate if the same receipt gets captured twice, and stores the image linked from the row — in a sheet the client owns. It is not practice software and does not try to be. For UK users it is worth noting the sheet itself is a record, not an MTD-compatible filing route; the filing happens through your accountant or compatible software.
If you are the bookkeeper reading this
The pragmatic position is that these are not really competitors for your use case. Dext is your production tooling. The relevant question is what you ask clients who refuse to use it to send you instead — and a structured sheet with linked receipt images is a far better answer than a folder of photographs, because it arrives already categorised and already evidenced.
Check Dext's current plans and pricing on their own site before deciding; practice software pricing changes and tiers differ by region.